检索结果-内蒙古大学图书馆

43rd AIAA DATC/IEEE Digital Avionics Systems Conference, DASC 2024

作者： Kim, Jiwon Wang, Kexin Dorneich, Michael Winer, Eliot Brown, Lori Whitehurst, Geoffrey Iowa State University Industrial and Manufacturing Systems Engineering AmesIA United States Iowa State University Human Computer Interaction AmesIA United States Iowa State University Mechanical Engineering AmesIA United States Western Michigan University Aviation Flight Science KalamazooMI United States

ISBN: (纸本)9798350349610

This paper presents the preliminary evaluation results of an extended reality (XR) authoring tool designed for aviation weather training. XR has the potential to improve educational outcomes in flight training programs. However, the complexity and time demands of existing XR authoring tools have slowed wider adoption. This work strives to address these gaps by developing and continuously refining an instructor-centered, user-friendly XR authoring tool. Seventeen flight instructors participated in developing two XR lessons and evaluated the tool's effectiveness. Results indicated that the introduction of the XR authoring tool significantly boosted the instructors' confidence in creating XR content and alleviated concerns related to time consumption. Additionally, the findings showed that the XR authoring tool is highly usable and demands a low workload, with the XR lessons produced being highly satisfactory. By employing an instructor-centered approach, the XR authoring tool enabled instructors who are not XR experts to improve students' learning experiences, focusing on aviation weather content without the burden of technical complexities. © 2024 IEEE.

关键词： Educational technology

来源：评论

学校读者我要写书评

暂无评论

Head-wise Shareable Attention for Large Language Models

Head-wise Shareable Attention for Large Language Models

引用

2024 Findings of the Association for Computational Linguistics, EMNLP 2024

作者： Cao, Zouying Yang, Yifei Zhao, Hai Department of Computer Science and Engineering Shanghai Jiao Tong University China Key Laboratory of Shanghai Education Commission for Intelligent Interaction and Cognitive Engineering Shanghai Jiao Tong University China Shanghai Key Laboratory of Trusted Data Circulation and Governance in Web3 China

ISBN: (纸本)9798891761681

Large Language Models (LLMs) suffer from huge number of parameters, which restricts their deployment on edge devices. Weight sharing is one promising solution that encourages weight reuse, effectively reducing memory usage with less performance drop. However, current weight sharing techniques primarily focus on small-scale models like BERT and employ coarse-grained sharing rules, e.g., layer-wise. This becomes limiting given the prevalence of LLMs and sharing an entire layer or block obviously diminishes the flexibility of weight sharing. In this paper, we present a perspective on head-wise shareable attention for large language models. We further propose two memory-efficient methods that share parameters across attention heads, with a specific focus on LLMs. Both of them use the same dynamic strategy to select the shared weight matrices. The first method directly reuses the pre-trained weights without retraining, denoted as DirectShare. The second method first post-trains with constraint on weight matrix similarity and then shares, denoted as PostShare. Experimental results reveal our head-wise shared models still maintain satisfactory capabilities, demonstrating the feasibility of fine-grained weight sharing applied to LLMs. © 2024 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

SIAMIXFORMER: A FULLY-TRANSFORMER SIAMESE NETWORK WITH TEMPORAL FUSION FOR ACCURATE BUILDING DETECTION AND CHANGE DETECTION IN BI-TEMPORAL REMOTE SENSING IMAGES

arXiv

引用

arXiv 2022年

作者： Mohammadian, Amir Ghaderi, Foad Human-Computer interaction lab Electrical and Computer Engineering Department Tarbiat Modares University Tehran Iran

Building detection and change detection using remote sensing images can help urban and rescue planning. Moreover, they can be used for building damage assessment after natural disasters. Currently, most of the existing models for building detection use only one image (pre-disaster image) to detect buildings. This is based on the idea that post-disaster images reduce the model’s performance because of presence of destroyed buildings. In this paper, we propose a siamese model, called SiamixFormer, which uses pre- and post-disaster images as input. Our model has two encoders and has a hierarchical transformer architecture. The output of each stage in both encoders is given to a temporal transformer for feature fusion in a way that query is generated from pre-disaster images and (key, value) is generated from post-disaster images. To this end, temporal features are also considered in feature fusion. Another advantage of using temporal transformers in feature fusion is that they can better maintain large receptive fields generated by transformer encoders compared with CNNs. Finally, the output of the temporal transformer is given to a simple MLP decoder at each stage. The SiamixFormer model is evaluated on xBD, and WHU datasets, for building detection and on LEVIR-CD and CDD datasets for change detection and could outperform the state-of-the-art. Copyright © 2022, The Authors. All rights reserved.

关键词： Disasters

来源：评论

学校读者我要写书评

暂无评论

Audio-Driven Facial Landmark Generation in Violin Performance using 3DCNN Network with Self Attention Model 48

Audio-Driven Facial Landmark Generation in Violin Performanc...

引用

48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023

作者： Lin, Ting-Wei Liu, Chao-Lin Su, Li Taiwan International Graduate Program Social Networks and Human-Centered Computing Taiwan Academia Sinica Institute of Information Science Taipei Taiwan National Chengchi University Department of Computer Science Taipei Taiwan

ISBN: (纸本)9781728163277

In a music scenario, both auditory and visual elements are essential to achieve an outstanding performance. Recent research has focused on the generation of body movements or fingering from audio in music performance. The audio-driven face generation technique in music performance is still deficient. In this paper, we compile a violin soundtrack and facial expression dataset (VSFE) for modeling facial expressions in violin performance. To our knowledge, this is the first dataset mapping the relationship between violin performance audio and musicians' facial expressions. We then propose a 3DCNN network with self-attention and residual blocks for audio-driven facial expression generation. In the experiments, we compare our methods with three baselines on talking face generation. The codes and dataset are available on the Github (https://***/kevinlin91/icassp-music2face). © 2023 IEEE.

关键词： Music

来源：评论

学校读者我要写书评

暂无评论

Research on the Ecological Environment Evaluation of Technological Talents in the Shuangcheng Economic Circle of Chengdu Region 1

引用

International Conference on Data Analytics in Business and Marketing, ICDABM 2022

作者： Liu, Hong Yadav, Amit Hu, Li Qin Mehedi Hassan, Md. Mollick, Swarnali Talwar, Dinesh Department of Human Resource Management Chengdu University of Technology Chengdu China Darwin Australia Department of E-Commerce Chengdu Neusoft University Chengdu China Computer Science and Engineering North Western University Khulna Bangladesh Computer Science and Engineering Northern University of Business and Technology Khulna Bangladesh Department of Corporate and Digital Development Darwin Australia

ISBN: (数字)9789819953585

ISBN: (纸本)9789819953578

This paper takes 22 cities in Sichuan-Chongqing region as examples, builds a comprehensive evaluation index system for the ecological environment of scientific and technological talents on the basis of existing research, uses the mean square error weighting method to determine the weight of each index, and uses the comparative analysis method to analyze the Chengdu-Chongqing double-city economic circle. The current situation and shortcomings of the cities in the country proposed countermeasures and suggestions to improve the comprehensive ability of scientific and technological talents in the ecological environment. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Mean square error

来源：评论

学校读者我要写书评

暂无评论

Leveraging Part-of-Speech Tagging for Enhanced Stylometry of Latin Literature 1

Leveraging Part-of-Speech Tagging for Enhanced Stylometry of...

引用

1st Workshop on Machine Learning for Ancient Languages, ML4AL 2024

作者： Chen, Sarah Li Burns, Patrick J. Bolt, Thomas J. Chaudhuri, Pramit Dexter, Joseph P. Phillips Academy Andover United States Research Science Institute Center for Excellence in Education United States Department of Computer Science Stanford University United States Institute for the Study of the Ancient World New York University United States Department of Languages & Literary Studies Lafayette College United States Department of Classics University of Texas Austin United States Institute of Collaborative Innovation Department of Computer and Information Science University of Macau China Department of Human Evolutionary Biology Harvard University United States

ISBN: (纸本)9798891761445

In literary critical applications, stylometry can benefit from hand-curated feature sets capturing various syntactic and rhetorical functions. For premodern languages, calculation of such features is hampered by a lack of computational resources for accurate part-of-speech tagging and semantic disambiguation. This paper reports an evaluation of POS taggers for Latin and their use in augmenting a hand-curated stylometric feature set. Our analyses show that POS-augmented features not only provide more accurate counts but also perform well on tasks such as genre classification. In the course of this work, we introduce POS n-grams as a feature for Latin stylometry. © 2024 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Domain-aware Multimodal Dialog Systems with Distribution-based User Characteristic Modeling

引用

ACM Transactions on Multimedia Computing, Communications and Applications 2024年第2期21卷 1-22页

作者： Chen, Xiaolin Song, Xuemeng Zuo, Jianhui Wei, Yinwei Nie, Liqiang Chua, Tat-Seng School of Software Joint SDU-NTU Centre for Artificial Intelligence Research Shandong University Jinan China School of Computer Science and Technology Shandong University Jinan China Department of Human Centred Computing Monash University Melbourne Australia Shenzhen China School of Computing National University of Singapore Singapore Singapore

Textual response generation is a pivotal yet challenging task for multimodal task-oriented dialog systems, which targets at generating the appropriate textual response given the multimodal context. Although existing efforts have obtained remarkable advancements, they ignore the potential of the domain information in revealing the key points of the user intention and the user's history dialogs in indicating the user's characteristics. To address this issue, in this work, we propose a novel domain-aware multimodal dialog system with distribution-based user characteristic modeling (named DMDU). In particular, DMDU contains three vital components: context-knowledge embedding extraction, domain-aware response generation, and distribution-based user characteristic injection. Specifically, the context-knowledge embedding extraction component aims to extract the embedding of multimodal context and related knowledge following existing studies. The domain-aware response generation component targets at conducting domain-aware fine-grained intention modeling based on the context and knowledge embedding, and thus fulfills the textual response generation. Moreover, the distribution-based user characteristic injection component first captures the user's characteristics and current intention with the Gaussian distribution and then conducts the sampling-based contrastive semantic regularization to promote the context representation learning. Experimental results on the public dataset demonstrate the effectiveness of DMDU. We release codes to promote other researchers. © 2024 Copyright held by the owner/author(s). Publication rights licensed to ACM.

关键词： Gaussian distribution

来源：评论

学校读者我要写书评

暂无评论

Designing a Sustainable Material for 3D Printing with Spent Coffee Grounds 23

Designing a Sustainable Material for 3D Printing with Spent ...

引用

Proceedings of the 2023 ACM Designing Interactive Systems Conference

作者： Michael L. Rivera S. Sandra Bae Scott E. Hudson ATLAS Institute and Department of Computer Science University of Colorado Boulder United States ATLAS Institute University of Colorado Boulder United States Human-Computer Interaction Institute Carnegie Mellon University United States

ISBN: (纸本)9781450398930

The widespread adoption of 3D printers exacerbates existing environmental challenges as these machines increase energy consumption, waste output, and the use of plastics. Material choice for 3D printing is tightly connected to these challenges, and as such researchers and designers are exploring sustainable alternatives. Building on these efforts, this work explores using spent coffee grounds as a sustainable material for prototyping with 3D printing. This material, in addition to being compostable and recyclable, can be easily made and printed at home. We describe the material in detail, including the process of making it from readily available ingredients, its material characteristics and its printing parameters. We then explore how it can support sustainable prototyping practices as well as HCI applications. In reflecting on our design process, we discuss challenges and opportunities for the HCI community to support sustainable prototyping and personal fabrication. We conclude with a set of design considerations for others to weigh when exploring sustainable materials for 3D printing and prototyping.

关键词： environmental sustainability zero-waste prototyping personal fabrication 3D printing bio-based materials

来源：评论

学校读者我要写书评

暂无评论

A Machine Learning Approaches for Natural Language Processing and Sentiment Analysis in Social Media 7

A Machine Learning Approaches for Natural Language Processin...

引用

7th International Conference on Contemporary Computing and Informatics, IC3I 2024

作者： Manoharan, Geetha Sarkar, Prithu Pathak, Punit Ahmad, Khadri Syed Faizz Singh, Arun Vaishnavi, M. Sr University School of Business Telangana Warangal India Amity University Amity School of Communication Kolkata India Auro University School of Liberal Arts and Human Sciences Surat India Srm University Department Computer Science and Engineering Andhra Pradesh India Vikrant University M.P Gwalior India Bannari Amman Institute of Technology Department of Computer Science and Engineering Erode India

ISBN: (纸本)9798350350067

In the recent decade, there has been a rise in social media and various companies are using it as a primary tool for marketing. In order to better understand the emotion and views of the customers, social media sentiment analysis has been used. Machine learning is a rising technology which helps in sentiment analysis in the social media which helps to easily categorize the views and opinions of the customers in certain categories. The major objectives of the study include exploring Machine Learning Approaches for Natural Language Processing and Sentiment Analysis in social media. In this research, the methods of sentiment analysis in online social networks using ML and techniques for grouping have been studied in detail. A secondary research approach has been followed for the research. The major benefits of sentiment analysis on social media have also been presented as a part of the report. In the research it has been found that there are two major types of clarifying sentiments which are supervised and unsupervised learning. The research also reveals that the best learning model for evaluating the Twitter dataset is determined to be CNN using Word to Vector and under instruction. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Exploring Effective Early Research Exposure for Broadening Participation in Computing science 2025

Exploring Effective Early Research Exposure for Broadening P...

引用

Proceedings of the 30th ACM Conference on Innovation and Technology in computer science Education V. 2

作者： Ouldooz Baghban Karimi Rebecca Robinson Shanon Reckinger Giulia Alberini Sruti Bhagavatula Trevor Bonjour Dimitrij (Mitja) Hmeljak Konstantinos Liaskos Susan H. Rodger Ruchi Sembey Megan Venn-Wycherley School of Computing Science Simon Fraser University Surrey BC Canada Department of Human Centred Computing Monash University Clayton Victoria Australia Computer Science University of Illinois Chicago Chicago IL USA School of Computer Science McGill University Montreal QC Canada Department of Computer Science Northwestern University Evanston IL USA Department of Computer Science and Engineering University of California San Diego San Diego CA USA Department of Computer Science Indiana University Bloomington IN USA Department of Computer Science School of Mathematical and Computer Sciences Heriot-Watt University Edinburgh Scotland United Kingdom Department of Computer Science Duke University Durham NC USA Department of Computer Science Swansea University Swansea United Kingdom

ISBN: (纸本)9798400715693

Evidence supports offering research experiences for undergraduate computing science students as a means of broadening participation in computing[7, 10, 11]. However, student perceptions about computing science research, how students become interested in these research experiences, and the details of effective design and delivery of programs capable of attracting and retaining this interest are less explored. In this study, we investigate the design and delivery of undergraduate research programs. We expand on and explore several factors, including but not limited to cultural relevance, the presence of a cross-disciplinary high-level view, task assignment, entry point, and support elements of a program.

关键词： broadening participation

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：